Can we identify reverberant cupped hands coloration with measurements?

The in context of what you quoted, the "reverberant cupped hands coloration", it would be regular ripples between 1.5kHz to 4kHz up to 2ms. Anything after 2ms or from narrow vertical ridges should not be considered as it's normal for ringing from peaks to exhibit this effect as it dies.

Refer to the three examples below. Note that the CSD floor for the Nighthawk and T1 Gen 2 have been lowered to better show this effect. The DT1770 is the worst offender among these headphones terms of the cavern effect.

 

This are my thoughts:

A reflection happens when a replica of what was produced bounces back into the receiver (or transmitter). This is some sort of an echo. In sound I guess some folks call it reverberation. Usually in these cases we are dealing with stuff that bounces back after a time which is larger than the time duration of an impulse response (or the majority of it's energy localization). That is, when we are talking reverberation. But that's not necessarily the end of the story.

When a reflection, or a large number of reflections (sometimes infinitesimal) happen well within the range of an impulse response, we have filtering (analog or digital).

And so, reflections with in range of the impulse response results in coloration. Cupped hands should result in short length reflections, possibly with in range of the impulse response. This results in standing waves which produces notches and peaks and stuff. Which in turn may also be visualized in CSD plots.

To me some of these issues sound like room coloration. Wonky. As in Madonna singing True Blue in the bathroom or Papa Don't Preach with her head stuck in the trash can. At other times, in cases perhaps where standing waves add up as a function of their phase, we can have a bunch of peaks that occasionally drill ones ears suddenly and unexpectedly.

 

No reverb-ry headphones for comparison.

 

Just a slight amount and different in nature. Not sure if this can be heard as "cup reverb" per the closed headphones.

 

One millisecond is approximately one foot. So makes sense in the above. If we see about five or six waves, that's about two inches from the eardrum or the ears to the edge of the cups. Seems about right.